graham/prop
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
prop/lib/microwaveprop/workers
History
Graham McIntire 15ce50365f
perf(concurrency): partition Task.Supervisor to remove bottleneck on heavy async_stream paths 
Wrap a Task.Supervisor in a PartitionSupervisor (one partition per
scheduler) and route the four sustained async_stream callers through
`{:via, PartitionSupervisor, {Microwaveprop.TaskSupervisor, self()}}`.
Concurrent work — the hourly PropagationGridWorker pulling f00-f18
HRRR range downloads, the GEFS worker scoring its grid, and the
Recalibrator's 20-way positive/negative fan-out — no longer funnels
through a single supervisor PID.

Light/one-shot Task.async + Task.start sites (hrrr_client surface/
pressure fan-out, application boot warmup, weather fire-and-forget)
are left alone; partitioning only helps under sustained concurrency.
2026-04-21 14:20:29 -05:00
..
admin_task_worker.ex fix(review): address code-reviewer findings 2026-04-21 10:09:46 -05:00
backfill_enqueue_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
canadian_sounding_fetch_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
common_volume_radar_worker.ex fix(workers): propagate transient errors so Oban retries + counters fire 2026-04-21 13:15:27 -05:00
contact_import_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
contact_position_backfill_worker.ex test: override stale distance_km when filling a missing pos 2026-04-16 14:09:45 -05:00
contact_weather_enqueue_worker.ex fix(dialyzer): actually fix LiveTable warnings instead of hiding them 2026-04-21 12:58:10 -05:00
gefs_fetch_worker.ex perf(concurrency): partition Task.Supervisor to remove bottleneck on heavy async_stream paths 2026-04-21 14:20:29 -05:00
grid_cache_prune_worker.ex feat(workers): add GridCachePruneWorker 2026-04-21 13:19:52 -05:00
hrrr_native_grid_worker.ex fix(dialyzer): actually fix LiveTable warnings instead of hiding them 2026-04-21 12:58:10 -05:00
iemre_fetch_worker.ex fix(workers): unique dedup on polling workers so queued dups don't stack 2026-04-21 14:12:43 -05:00
ionosphere_fetch_worker.ex Ionosphere: GIRO ionosonde ingestion (foF2/foEs/hmF2/MUFD) 2026-04-15 14:37:43 -05:00
mechanism_classify_worker.ex fix(workers): propagate transient errors so Oban retries + counters fire 2026-04-21 13:15:27 -05:00
narr_fetch_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
nexrad_worker.ex fix(dialyzer): actually fix LiveTable warnings instead of hiding them 2026-04-21 12:58:10 -05:00
propagation_grid_worker.ex feat(observability): instrument PropagationGridWorker with Instrument.span 2026-04-21 13:29:09 -05:00
propagation_prune_worker.ex fix(propagation): hourly grid chain preempts MRMS/prune backlog 2026-04-19 08:42:22 -05:00
radar_frame_worker.ex fix(workers): 3 bug/perf fixes from codebase review 2026-04-21 09:53:13 -05:00
rtma_fetch_worker.ex fix(workers): unique dedup on polling workers so queued dups don't stack 2026-04-21 14:12:43 -05:00
solar_index_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
space_weather_fetch_worker.ex SpaceWeather: SWPC JSON ingestion (Kp, F10.7, GOES X-ray) 2026-04-15 14:55:24 -05:00
terrain_profile_worker.ex fix(dialyzer): clear 125+ warnings under strict flags 2026-04-21 10:30:06 -05:00
weather_fetch_worker.ex perf(weather): batch ASOS observation upserts into a single insert_all 2026-04-21 13:19:53 -05:00